HoustonPutman opened a new issue #365: URL: https://github.com/apache/solr-operator/issues/365
Currently a SolrCloud is treated the same during a managed upgrade, whether it is using ephemeral or persistent data. However since all ephemeral data is lost during a pod update, the managed update logic fails to work when the ephemeral SolrCloud contains data. This is because the Solr Operator expects, after upgrading a pod, that the cluster state gets healthy before it can move on to the next pod. When the Solr replica data is deleted on pod upgrade, the collections those replicas belong to can never become healthy, since those replicas no longer exist. What we can do is migrate all replicas on a node before upgrading it, if the SolrCloud is using ephemeral data. We control when pods are upgraded so we can make sure that the pod contains no replicas before a delete occurs. If the ephemeral SolrCloud only contains 1 pod, then we have no choice to delete all the replicas for the pod before doing the upgrade, because there are no other pods to move them to. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
